engleski

A Reliability Improvement Method for SOA-Based Applications

As SOA gains more traction through various implementations, building reliable service compositions remains one of the principal research concerns. Widely researched reliability assurance strategies, like dynamic service compositions, often rely on spatial redundancy which can make them less applicable on a larger scale when discovering and invoking functionally equivalent services is a resource demanding task. To address this issue, we propose a design-time reliability improvement method. The proposed method enables selective composition improvements by utilizing the reliability model based on belief networks to recommend atomic services that have the greatest effect on the overall reliability. Since both the size of composition's reliability model and the resources required to discover service alternates affect the scalability of the proposed method, we introduce a suit of weak point recommendation algorithms with varying accuracy and complexity. Although less accurate and computationally lighter recommendation algorithms on average require more improvement steps, they can achieve better overall performance in cases when the overhead of detecting and applying service replacements per improvement step is low. We confirm the soundness of the proposed solution by performing experiments on data sets consistent of randomly generated service compositions taking into account varying service alternate discovery time.

Software reliability; modeling and prediction; web-based services

nije evidentirano